Qwen3-Coder-30B-A3B-Instruct: benchmark scores, price and ranking

Qwen3-Coder-30B-A3B-Instruct from Qwen ranks #53 of 743 LLMs on the LLMs Tiger aggregate benchmark score (52.7), based on 2 benchmark results captured up to 2026-10-13.

Organisation: Qwen
Type: Open-weight / local model
Open weights: Yes
Released: 2025-07-31 (source: curated)
Cheapest API price: $0.070 in / $0.270 out per million tokens (via novita, observed 2026-10-13T20:47Z)
Aggregate rank: #53 of 743 (score 52.7)

Across the 2 benchmarks where at least ten models are scored, Qwen3-Coder-30B-A3B-Instruct is in the top 10% on 1 and in the top half on 2.

Its cheapest listed input price, $0.070 per million tokens, is lower than that of 87% of the 427 models with a listed price.

Qwen3-Coder-30B-A3B-Instruct benchmark results

Qwen3-Coder-30B-A3B-Instruct benchmark scores, ranked against all models measured on each benchmark
Benchmark	Score	Rank	Source	Captured
SWE-bench Lite	49.7%	#2 of 20	SWE-bench Lite leaderboard	2026-10-13
SWE-bench Verified	60.4%	#33 of 73	SWE-bench Verified leaderboard	2026-10-13

About Qwen3-Coder-30B-A3B-Instruct

Qwen3-Coder-30B-A3B-Instruct is an open-weight model that can run on your own hardware from Qwen. It was released on 2025-07-31, according to the developer's announcement. The cheapest API listing we track charges $0.070 per million input tokens and $0.270 per million output tokens, through novita. We hold 2 benchmark results for it across 1 area, and 1 is in the top tenth of its board.

Where Qwen3-Coder-30B-A3B-Instruct stands

On our aggregate score, Qwen3-Coder-30B-A3B-Instruct is #53 of 743 ranked models, in the top quarter. Its score is 52.7; the scale is centred on 50, so it sits 2.7 points above the typical model. That beats 93% of the ranked field. Among the 23 ranked Qwen models it is #2. The next one up is Qwen3.5-122B-A10B at 53.0. The next one down is Qwen3.8-2.4T-A95B at 52.1.

Qwen3-Coder-30B-A3B-Instruct, benchmark by benchmark

On SWE-bench Lite, Qwen3-Coder-30B-A3B-Instruct scores 49.7% and ranks #2 of 20. Only one model scores higher, led by Claude 4 Sonnet at 56.7%, which is 7.0 points ahead. It covers real GitHub issues resolved. The board median is 24.7%, so Qwen3-Coder-30B-A3B-Instruct is 25.0 points above it. Source: SWE-bench Lite leaderboard, captured 2026-10-13.

On SWE-bench Verified, Qwen3-Coder-30B-A3B-Instruct reaches 60.4%. It ranks #33 of 73, ahead of 55% of the field, and 23.1 points behind Claude Opus 4.7 (max) at 83.5%. The results either side are claude-3-7-sonnet-20250219 (61.0%) and DeepSeek V3.2 Reasoner (60.0%). It covers real GitHub issues resolved. The board median is 56.4%, so Qwen3-Coder-30B-A3B-Instruct is 4.0 points above it. Source: SWE-bench Verified leaderboard, captured 2026-10-13.

Strengths and weaknesses

Software engineering: solid, ahead of 72% of the field on average across 2 benchmarks. Its best board here is SWE-bench Lite (#2 of 20) and its weakest is SWE-bench Verified (#33 of 73).

Price and value

The median input price across the 427 priced models is $0.500 per million tokens, so Qwen3-Coder-30B-A3B-Instruct costs less than the typical listing. Models that score within 1.5 points or better and cost less: DeepSeek V4 Flash 0731 (max) ($0.016, 53.1), GLM-5.2 (high) ($0.060, 51.6). Similar scores cost more on GLM-5.3 (max) ($0.079) and Hy3 ($0.083).

Release history

The Qwen model released before it was Qwen2.5 Coder 32B Instruct on 2024-11-12, which scores 48.3. The next one was qwen3-4b-instruct-2507 on 2025-08-06, at 48.7.

Reading these numbers

The 2 scores on this page come from 2 sources: SWE-bench Lite leaderboard (1), SWE-bench Verified leaderboard (1). No official leaderboard supplied any of them, so treat them as third-party measurements. With 2 results the aggregate rank can move a lot when one new score arrives.

Models ranked nearby

#50 Kimi K2.5 (Fireworks) (52.8)
#51 CodeAct v2.1 (claude-3-5-sonnet-20241022) (52.8)
#52 claude-3.5-sonnet-20240620 (52.7)
#54 GPT-5.1 (high) (52.7)
#55 Claude Sonnet 5 (max) (52.7)
#56 Claude 4.6 Opus (52.6)